English

Reward-Predictive Clustering

Machine Learning 2022-11-08 v1 Artificial Intelligence

Abstract

Recent advances in reinforcement-learning research have demonstrated impressive results in building algorithms that can out-perform humans in complex tasks. Nevertheless, creating reinforcement-learning systems that can build abstractions of their experience to accelerate learning in new contexts still remains an active area of research. Previous work showed that reward-predictive state abstractions fulfill this goal, but have only be applied to tabular settings. Here, we provide a clustering algorithm that enables the application of such state abstractions to deep learning settings, providing compressed representations of an agent's inputs that preserve the ability to predict sequences of reward. A convergence theorem and simulations show that the resulting reward-predictive deep network maximally compresses the agent's inputs, significantly speeding up learning in high dimensional visual control tasks. Furthermore, we present different generalization experiments and analyze under which conditions a pre-trained reward-predictive representation network can be re-used without re-training to accelerate learning -- a form of systematic out-of-distribution transfer.

Keywords

Cite

@article{arxiv.2211.03281,
  title  = {Reward-Predictive Clustering},
  author = {Lucas Lehnert and Michael J. Frank and Michael L. Littman},
  journal= {arXiv preprint arXiv:2211.03281},
  year   = {2022}
}
R2 v1 2026-06-28T05:17:55.087Z